1. Field of the Inventions
The present invention relates to displaying graphical output of clients in a client-server network, and to troubleshooting and maintaining the clients by an administrator.
2. Description of the Related Art
A “workstation blade” is a rack-mountable computer accessible over a network by a user from a remote user terminal networked to the workstation blade. A workstation blade is typically configured for use by a single user, replacing a local desktop computer workstation with a centrally located, rack-mountable computer. Computing resources such as the CPU, motherboard, hard drive, and videocards, reside on the workstation blade. The remote user terminal provides a familiar computing environment, including traditional peripherals, such as one or more user displays, a keyboard, and a mouse, even though these peripherals communicate over the Ethernet with the computing resources that are remotely located on the workstation blade. This consolidates computing resources for multiple users in a centralized location, allowing users to remotely interface with the centralized computing resources, while allowing a system administrator to more easily manage the computing resources. A system administrator terminal may be used to monitor applications of selected workstation blades over the network, such as to troubleshoot the workstation blade, install software, or perform other management tasks. The BLADECENTER product family from IBM includes some of the more compact chassis arrangements available on the market, and may be configured for use either with multi-user servers or with single-user workstation blades (IBM and BLADECENTER are registered trademarks of International Business Machines Corporation, Armonk, N.Y.). However, other rack-mountable workstation blade systems and their equivalents are available in the industry from a variety of manufacturers.
Workstation blades may be viewed in the context of client-server architecture, even though a server is often configured for use by multiple users and a workstation blade is typically configured for use by a single user. Client-server architecture is a computer network architecture wherein clients rely to some degree on the servers to which they are networked for processing. Each instance of the client software in a client-server network can send data requests to one or more connected servers. In turn, the servers can accept these requests, process them, and return the requested information to the client. As the concept of client-server architecture is applied to workstation blades, each instance of a workstation blade and associated user terminal form a client-server pair, wherein the workstation blade is on the server side and the user terminal is on the client side. The degree to which processing activities are performed on the server side of a client-server network varies among different systems and manufacturers. The term “thin client” (or “lean client”) is sometimes used to describe a client that has little processing capabilities and relies heavily on its associated server for processing. A thin client may be as simple as a “dumb-terminal” that has no processing capabilities other than conveying input and output between the user and the server, relying extensively or entirely on the computing capabilities of the server to which it is networked. Other types of clients have more processing power, and can run complete operating systems, but may still lack other components usually available on standalone computers, such as media drives.
When monitoring the graphical output (i.e. video output) of a workstation blade, it is helpful for the administrator terminal to display the same thing that is being displayed on the remote user display by the workstation blade. Monitoring a selected workstation blade from an administrator workstation can be problematic, however, due to potential differences in display capabilities and formats between the administrator display and the numerous individual user displays being used at the client terminals. Most notably, it can be difficult or impossible for an administrator to view the full graphical output of a user who has a multi-monitor setup having two or more physical display devices (e.g. CRT or LCD display), because the graphical (i.e. video) output of the user's workstation is split among the multiple display devices. This “split display” of the graphical output cannot be readily viewed by the administrator on the administrator's display terminal. Furthermore, each user's workstation may have different display drivers and display settings. This further complicates the problem of monitoring the user from the administrator workstation. A still further complication is that video signals from the various workstation blades are typically received and processed by a rack-mounted management module before being sent to the administrator terminal, and the management module has limited capacity for handling video signals. For example, the management module in many systems is limited to receiving a maximum 1024×768 display resolution, while the various workstation blades have graphics cards and associated hardware capable of much higher digital resolution.
A solution is therefore needed to allow a system administrator to remotely monitor the video output of a selected user workstation. In particular, a monitoring system is needed that would allow the administrator to monitor the workstation of any user regardless of the user display setup, including whether or not the user has a single monitor or multiple monitor display. The system administrator should be able to fully monitor the video from the administrator terminal in the same format as it would be displayed at the associated user terminal. The system administrator should also be able to monitor the video output regardless of the type, size, number, resolution, and other display parameters of the user terminal.